From: Greg Stark <gsstark@mit.edu>
To: Joseph Shraibman <jks@selectacast.net>
Cc: J. Andrew Rogers <jrogers@neopolitan.com>
Cc: pgsql-performance@postgresql.org
Subject: Re: linux distro for better pg performance
Date: 04 May 2004 02:03:16 -0400
Message-ID: <87k6zszq3f.fsf@stark.xeocode.com> (raw)
In-Reply-To: <40970058.3040608@selectacast.net>
References: <407E25A6.ADFE4815@t1.unisoftbg.com>
<407E909F.9000101@ehpg.net>
<1082053706.10823.38.camel@localhost.localdomain>
<40970058.3040608@selectacast.net>
Joseph Shraibman <jks@selectacast.net> writes:
> J. Andrew Rogers wrote:
>
> > Do these features make a difference? Far more than you would imagine. On one
> > postgres server I just upgraded, we went from a 3Ware 8x7200-RPM
> > RAID-10 configuration to an LSI 320-2 SCSI 3x10k RAID-5, with 256M
>
> Is raid 5 much faster than raid 10? On a 4 disk array with 3 data disks and 1
> parity disk, you have to write 4/3rds the original data, while on raid 10 you
> have to write 2 times the original data, so logically raid 5 should be faster.
In RAID5 every write needs to update the parity disk as well. In order to do
that for a small random access write you often need read in the rest of the
data block being modified to calculate the parity bits. This means writes
often have higher latency on RAID5 because they first have to do an extra
read. This is where RAID5 got its bad reputation.
Good modern RAID5 controllers can minimize this problem but I think it's still
an issue for a lot of lower end hardware. I wonder if postgres's architecture
might minimize it already just because of the pattern of writes it generates.
--
greg
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Reply to all the recipients using the --to and --cc options:
reply via email
To: pgsql-performance@postgresql.org
Cc: gsstark@mit.edu, jks@selectacast.net, jrogers@neopolitan.com
Subject: Re: linux distro for better pg performance
In-Reply-To: <87k6zszq3f.fsf@stark.xeocode.com>
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
This inbox is served by DDX for PostgreSQL; see mirroring instructions
for how to clone and mirror all data and code used for this inbox